Cognition and survival: an exploration in a large multicentre study of the population aged 65 years and over

Understanding the patterns in determinants of survival becomes increasingly important as the population ages. Dementia is known to shorten survival as is impaired cognition. Whether this is a continuous phenomenon and independent of other explanatory variables is less clear.

To examine a population-based dataset in which a measure of cognitive function (Mini-Mental State Examination [MMSE]), self-reported physical health and lifestyle variables were measured at outset, with monitoring for mortality thereafter.

The five identical sites of the Medical Research Council Cognitive Function and Ageing Study (MRC CFAS) were analysed, with populations in rural Cambridgeshire, Gwynedd, Newcastle, Nottingham and Oxford. Survival curves were modelled and stratified analyses carried out, with physical disease, sociodemographic variables and lifestyle variables as covariates.

There was a strong and consistent reduction in survival probability for each decrement in MMSE. Adjustment for known confounders did not alter this pattern. Social class and education in particular had no additional effect. Self-reported health was the only other associated variable.

Cognitive function appears to be a marker of capacity for survival in the UK. Terminal decline can account for some of this. Actuarial survival provided here can give carers and service providers an idea of prognosis at given ages and levels of cognition, and provide baseline data for those planning interventions in similar groups.